I am using a lock in amplifier detection technique along with an Helmholtz coil. I am measuring the real and imaginary part of the FMR signal.
Which signal (Amplitude, Real or imaginary) should be used while fitting and extracting the parameters? Amplitude signal is having completely different line shapes and kinks. While Real and imaginary part of the signal are completely smooth.
My suggestion is: first, locate where the peak is; then, auto phase at that position (field). This will make X=R (real equal to amplitude). And use the (differential) Lorentzian to fit to the real component.
Guanzhong Wu Do i need to do this at each frequency?
I am using the differential lorentzian fitting for Real part but amplitude is not lorentzian and it has some arbitrary lineshape with anomalous kinks and irregular line shape.
Vinay Sharma Could you show us one spectra to see which kind of feature you are talking ?
Bersweiler Mathias I am attaching the FMR (Real, Imaginary and amplitude) image @ 10GHz. You can see that Real and imaginary part of the signal are completely smooth, but amplitude signal is not showing any defined profile.
Vinay Sharma Thanks a lot.
I guess your amplitude is the superposition of both signal right ?
Generally what we do, is to fit the real part or the imaginary part using a superposition of a Lorentzian + antiLorentzian + linear function to the spectrum, from which resonance frequency and linewidth of each spectrum are determined. The linear function is added to remove the residual background signal after subtracting the experimentally determined background.
You can also try a combined fit (Lorentz+ anti Lorentz) of the real and imaginary part using shared parameters (resonance frequency and lindwidth), you should get almost the same result as for a single fit of the imaginary or real part.
From your plot, the imaginary curve looks nice, however I didn't expect the same shape for the real part. I expected something like in the pic (see attached file but it is in case of VNA-FMR). So maybe check how you determin your real and imaginary part too (just to be sur your real part is correct).
Bersweiler Mathias Thank you for your response.
Yes the amplitude is calculated by the form (x+iy) using the eulers form.
However i just want to know that why the amplitude looks strange? I am measuring in differential form (Amplitude and phase of the signal) using Lock in detection technique.
I am also doing the same thing fitting my real and imaginary curves using symmetric and antisymmetric component of lorentzian fitting equation and calculating the FMR linewdith and resonance point.
But can we fit the amplitude also with this function?
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts